RBX vs WW
 

Been waiting on this for a while , down to the purchasing time , cant decide, not much in price difference, and from the reviews sometimes the RBX comes out on top and sometimes the WW, read some post where cathar says WW takes off at higher flow.
But also seen some say the same for RBX.

Now my system is 1048 chevette heater core and a dangerden gf4/radeon 9700 gpu block, Does anyone know what the average flow is for that setup and which block would be best, Im only posting because i myself cannot decide and i need more input then what i am finding on the net.

The 1048 isn't going to cut it for either block. If your set on that pump get a high flow rate block like the DD Maze 4 or D-Tek TC-4.

Not set on the pump just no money for a new one, the reviews ive seen, among other thigns have used this pump and C/W wasnt bad at all would it be temporarly decent?

My system consists of a White Water, an MCW50, a large (somewhat more restrictive than typical) heatercore, four 3/4" ball valves, about 10 feet of 1/2" ID tubing and about 10 feet of 5/8" ID tubing.

According to my mobo's temp sensor, my temp dropped ~1C when going from a 1048 to an Iwaki MD20-R.

I'd say the 1048 will do just fine.

Specify the review using a C/W value and that pump? Also remember you have a GPU block which will cut the flow rate and pressure significantly. I wouldn't even consider either block with that setup. Sure it will work though. Just don't expect great temps.

Read http://www.overclockers.com/articles886/ one of the few that can take a decent C/W value.

So will buying $25 water block instead. ;)

im looking more toward the future im probaly going to get a bigger pump , im was reviewing to reading some of cathars post on the overclockers forums where he listed he was talking about the 1048

If you want a good low flow block, the new swifty mcx5002 (I think its called) would be good.

If upgrading to a new pump, go ahead and get a mag 3, 350gph, 10.5ft head pressure, and relatively cheap and very reliable.

Cathar has previously commented that with a WW and an average heatercore in a system, the 1048 will perform quite well. I can't be arsed searching for the exact results now, but the 1250 only gained like 1º over the 1048 with the WW in this sort of situation.

However as soon as you add extra blocks you are pushing the limit.

The 1048 will do fine for now, but expect it to be a few degrees below top performance.

I've go the WW, simply because I think the RBX design is flawed. It is just trying to sell blocks on idea that adding two together gives better performance. It doesn't necessarily work.

eh, its the MCW5002, not MCX, the MCX range is aircooled

Sure untill you add a restrictive block in the loop like "a dangerden gf4/radeon 9700 gpu block" as he said. Does Cathar test with addded restrictions?

Anyway just do it. Should work ok.

It will be fine. If you get an RBX, you can adjust it for more flow anyway, though I doubt anything would be gained by it. All of these watercoolers are all about throwing money at things. I'd bet that either of those blocks would outperform the "low flow" blocks in your current configuration and rock balls if you upgraded the pump.

Presumably DD started with a traditional WW design and tweaked it until it performed better than the existing blocks. Wouldn't be much reason to make changes otherwise. I think it would've been nice if DD had tipped the hat as to the inspiration of the block (would have been nice for the slit edge too eh?) but honestly their design is at least as different from the WW as many many blocks were from the newer revision Maze 2s. That's life. The price that they can sell the RBX for is impressive to me; I think it's an example of DD doing what they have always done: keep the price of good wc performance down low for average buyers.

It differs from the WW only where it in the same as the Cascade.

I have some other issues with the block though:
It doesn't appear to cover the AMD mounting pads, whoich means a higher chance of crushing your core, especially just from slightly movement and hose tugging down on one side.
The nozzle system. I mean 6 nozzles? Why not just supply the best one instead of making people screw around. From the reviews I have seen it was only one nozzle that provided WW level cooling, and this nozzle wasn't supplied as standard, only in the 6 nozzle kit.

It may be a really good block, but I'm not so sure.

It's really not that similar to a cascade at all. If you look at what makes the cascade WORK then it's completely different.

that depends, what if you use the cascade type nozzle plate?

they have one that is like many straws sticking down into cups? i assumed it was merely holes...

It is nothing like the RBX in any way. The cups in the RBX only purpose is to create extra surface area, they are not used for jet impingement. It is nothing more than a White Water with wavy fins, some shallow cups, wider and less channels, and some changeable jets.

Also DD is shipping extra pads to use in place of the AMD pads that it doesn't touch, also the RBX does not cover the entire surface of a IHS.

I thought we covered this already.

Yep! !;)

Oh, now I remember :D at least I got the numbers right...

Errrr The cascade is more restrictive than both the WW and the RBX, and cathar said mine would be fine if i added a MCW50 into the equation (Cascade, LA heatercore, MCW50, RES, 1048)

As long as its got something like 4LPH going through it, it should be ok....

People and 1048 bashing :rolleyes:

What? 4LPH is only 1 gallon per HOUR. You must mean mean 4LPM which is 60 gallon per hour. The pump is only rated at 158GPH. Whatever.

Well, the reasons are these, from my perspective.

The RBX looks to me like its channels are cut with a 1/16" cutter (~1.59mm). This allows for dramatically faster machining (~2.5x) than the White Water with its 1.0mm channels/cutter.

Due to the obvious loss in fin/channel density, the only way to claw back some lost performance is by introducing some turbulence mechanisms, such as wavy fin walls and a few dips (voids) here and there, which are fairly "low cost" to machine in terms of time needed.

So basically the RBX looks like it'd probably take about half the machine time of the White Water to carve it out. From a machine-time/performance/cost perspective, this then makes good sense as to why it's made the way it is. The RBX would need about half the time on a milling machine as the White Water does.

However, the straight-forward nature of the White Water means that it doesn't necessarily need to be made on a milling machine.

Basically it's all about making the best trade-off between machine time/effort/cost by attempting to cheaply make up for some of what was lost.

From that approach, it is a fairly unique adaptation of the White Water design.

It is absolutely and positively nothing like a Cascade though.